Cocaine dependence is a major public health problem and the development of a treatment for this disorder is a priority. To date, treatment interventions based on reinforcement principles have shown the most robust empirical support for improving substance use outcomes. One example is contingency management (CM) interventions in which nondrug alternatives are introduced into the environment with the goal of increasing the ratio of reinforcement derived from nondrug stimuli to the ratio of drug-derived reinforcement. Cocaine dependent individuals vary in their responsiveness or sensitivity to nondrug reinforcers, however, which has been a limiting factor in the efficacy of CM interventions. Recent lines of evidence suggest that certain medications add efficacy to CM interventions, particularly agents that target dopaminergic (DA) actions in cocaine reinforcement. A particularly promising medication that we have investigated is the DA precursor levodopa, an agent that increases the store of available central DA in DA-deficient individuals. We have observed the strongest effects of levodopa when the medication is administered in the context of CM therapy. A plausible explanation for the observed efficacy is that levodopa, when given concomitantly with CM, enhances the saliency value of the incentives, which in turn motivates goal-directed behavior (abstinence). The proposed study is designed to replicate, predict, and extend the effects of this promising treatment approach. Cocaine dependent outpatients will participate in a randomized, 2-group (levodopa vs. placebo), double-blind clinical trial. Abstinent-based CM procedures will serve as the behavioral therapy platform. Potential predictors related to attentional bias (cocaine Stroop task) and motivation (modified progressive-ratio procedure) will be assessed at baseline and during treatment. We hypothesize that CM+levodopa will be more effective than CM+placebo in reducing cocaine use and that the effects of CM+levodopa will be associated with levels of attentional bias and motivation before and during treatment. A secondary aim will explore whether levodopa therapy contributes to the persistence of abstinence after discontinuation of 'CM. In conclusion, this study will validate the usefulness of a novel behavioral-pharmacological treatment and shed light on the mechanisms underlying the synergism between DA augmentation strategies and reinforcement-based therapies.